Mirror/stable-diffusion-webui
1
0
Fork 0
mirror of https://github.com/AUTOMATIC1111/stable-diffusion-webui.git synced 2024-06-07 21:20:49 +00:00
Code Issues Packages Projects Releases Wiki Activity

Fix tensorboard related functions

Browse Source
This commit is contained in:
aria1th 2023-01-16 03:02:54 +09:00
parent 598f7fcd84
commit 13445738d9
1 changed files with 6 additions and 7 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
modules/hypernetworks/hypernetwork.py
View File

@ -561,7 +561,7 @@ def train_hypernetwork(id_task, hypernetwork_name, learn_rate, batch_size, gradi
_loss_step = 0 #internal _loss_step = 0 #internal
# size = len(ds.indexes) # size = len(ds.indexes)
# loss_dict = defaultdict(lambda : deque(maxlen = 1024)) # loss_dict = defaultdict(lambda : deque(maxlen = 1024))
loss_logging = [] loss_logging = deque(maxlen=len(ds) * 3) # this should be configurable parameter, this is 3 * epoch(dataset size)
# losses = torch.zeros((size,)) # losses = torch.zeros((size,))
# previous_mean_losses = [0] # previous_mean_losses = [0]
# previous_mean_loss = 0 # previous_mean_loss = 0
@ -602,7 +602,6 @@ def train_hypernetwork(id_task, hypernetwork_name, learn_rate, batch_size, gradi
else: else:
c = stack_conds(batch.cond).to(devices.device, non_blocking=pin_memory) c = stack_conds(batch.cond).to(devices.device, non_blocking=pin_memory)
loss = shared.sd_model(x, c)[0] / gradient_step loss = shared.sd_model(x, c)[0] / gradient_step
loss_logging.append(loss.item())
del x del x
del c del c
@ -612,7 +611,7 @@ def train_hypernetwork(id_task, hypernetwork_name, learn_rate, batch_size, gradi
# go back until we reach gradient accumulation steps # go back until we reach gradient accumulation steps
if (j + 1) % gradient_step != 0: if (j + 1) % gradient_step != 0:
continue continue
loss_logging.append(_loss_step)
if clip_grad: if clip_grad:
clip_grad(weights, clip_grad_sched.learn_rate) clip_grad(weights, clip_grad_sched.learn_rate)
@ -691,9 +690,6 @@ def train_hypernetwork(id_task, hypernetwork_name, learn_rate, batch_size, gradi
processed = processing.process_images(p) processed = processing.process_images(p)
image = processed.images[0] if len(processed.images) > 0 else None image = processed.images[0] if len(processed.images) > 0 else None
if shared.opts.training_enable_tensorboard and shared.opts.training_tensorboard_save_images:
textual_inversion.tensorboard_add_image(tensorboard_writer, f"Validation at epoch {epoch_num}", image, hypernetwork.step)
if unload: if unload:
shared.sd_model.cond_stage_model.to(devices.cpu) shared.sd_model.cond_stage_model.to(devices.cpu)
shared.sd_model.first_stage_model.to(devices.cpu) shared.sd_model.first_stage_model.to(devices.cpu)
@ -703,7 +699,10 @@ def train_hypernetwork(id_task, hypernetwork_name, learn_rate, batch_size, gradi
hypernetwork.train() hypernetwork.train()
if image is not None: if image is not None:
shared.state.assign_current_image(image) shared.state.assign_current_image(image)
if shared.opts.training_enable_tensorboard and shared.opts.training_tensorboard_save_images:
textual_inversion.tensorboard_add_image(tensorboard_writer,
f"Validation at epoch {epoch_num}", image,
hypernetwork.step)
last_saved_image, last_text_info = images.save_image(image, images_dir, "", p.seed, p.prompt, shared.opts.samples_format, processed.infotexts[0], p=p, forced_filename=forced_filename, save_to_dirs=False) last_saved_image, last_text_info = images.save_image(image, images_dir, "", p.seed, p.prompt, shared.opts.samples_format, processed.infotexts[0], p=p, forced_filename=forced_filename, save_to_dirs=False)
last_saved_image += f", prompt: {preview_text}" last_saved_image += f", prompt: {preview_text}"